Face liveness detection for combating the spoofing attack in face recognition

The spoofing attack, in which the attacker disguises himself or herself as an authenticated user by showing the user's photographs or video clips before the camera, is a common adversarial attack in face recognition. To ensure the security of the system, face liveness detection is carried out to distinguish the images captured from a live face from those from a forged face. In this paper, we propose a face liveness detection method based on the High Frequency Descriptor to combat the spoofing attack. The additional illumination is added, which can raise the energy of high frequency components of a real face by exposing more details of the hair and skin, and lower it by causing a glisten on the planar surface. The difference of the energy of high frequency components between images with and without the illumination is calculated. Experimental results show that our method outperforms the original method and has robustness when the resolution of the attack media is high.